Tordivel AS today announced that it will present Scorpion Vision AI at VISION 2026, the world's leading trade fair for machine vision, taking place at Messe Stuttgart from 6–8 October 2026.
Scorpion Vision AI has been operational since Easter 2026 and remains in continuous development. It extends Tordivel's established Scorpion Vision platform with an interface designed for collaboration between machine-vision engineers and compatible AI agents.
Through the Scorpion Model Context Protocol (MCP) interface, AI can inspect and interact with a configured vision application, including cameras, images, calibration, tools, parameters, results and execution sequences. The interface currently exposes more than 400 Scorpion methods.
Scorpion's integrated Python environment allows reusable components to combine proven industrial vision functionality with approved open-source technologies such as image-processing, AI inference, 3D-processing, communication and data-analysis libraries. AI can assist engineers in creating components, configuring applications, testing against representative images, analysing results and documenting changes.
Importantly, the deployed production application remains controlled and deterministic. AI supports engineering and maintenance, while approved and versioned components execute within the established Scorpion runtime without requiring a generative AI service to make production decisions.

“Python has been an integral part of Scorpion for more than two decades. By combining this foundation with our MCP interface, Scorpion Vision AI gives engineers a practical way to use AI and the open-source ecosystem inside a professional industrial vision platform,” said Thor Vollset, Managing Director of Tordivel AS. “The objective is not to replace deterministic machine vision. It is to make the development, testing, deployment and long-term maintenance of advanced vision systems substantially faster and more accessible.”

From AI assistance to deployable industrial components

Scorpion Vision AI is designed to support the complete application lifecycle:

• Inspect existing Scorpion profiles and application structures.
• Configure cameras, calibration, 2D and 3D vision tools and communication.
• Create reusable Python-based vision components.
• Use approved packages from the wider Python and open-source ecosystem.
• Test components against stored or live production images.
• Compare results, performance and execution time.
• Document, version and deploy validated application changes.
• Diagnose and maintain installed systems through a controlled interface.

The platform builds on Scorpion's long-established capabilities in configurable 2D and 3D machine vision, robot guidance, precision measurement, inspection, colour processing, AI inference and industrial communication. It is intended for OEM machine builders, system integrators and manufacturers requiring maintainable vision systems that can evolve over long machine lifecycles.

Meet Tordivel and Scorpion Vision at VISION 2026

Visitors to VISION 2026 will be able to discuss Scorpion Vision AI, Python component development, MCP-based AI integration, OEM machine-vision platforms and applications in 2D vision, 3D scanning, robot vision, food inspection and logistics.

Event: VISION 2026
Date: 6–8 October 2026
Location: Messe Stuttgart, Stuttgart, Germany
Stand: 10H75
website: https://www.messe-stuttgart.de/vision/en/

Download Press release

‍